Article Text

Download PDFPDF

Association between hypoparathyroidism and defective T cell immunity in 22q11.2 deletion syndrome
  1. A Herwadkar1,
  2. A R Gennery2,
  3. A S Moran1,
  4. M R Haeney1,
  5. P D Arkwright3
  1. 1Department of Immunology, Salford Royal Hospital NHS Trust, Manchester, UK
  2. 2University of Newcastle, Newcastle General Hospital, Newcastle upon Tyne, UK
  3. 3University of Manchester, Royal Manchester Children's Hospital, Manchester, UK
  1. Correspondence to Dr P D Arkwright, University of Manchester, Royal Manchester Children's Hospital, Oxford Road, Manchester M13 9WL, UK; peter_arkwright{at}


  • Funding No external funding.

  • Competing interests None.

  • Provenance and peer review Not commissioned; externally peer reviewed.

View Full Text

Statistics from

The 22q11.2 chromosome region consists of at least 35 genes, some of which are important in the development of fourth branchial arch artery structures (including the great vessels of the heart), and third and fourth pharyngeal pouch structures (including the thymus and parathyroid glands). Other genes in this region are important in the development of palatal and facial structures. Thymic hypoplasia leading to defective T lymphocyte maturation is a feature of 22q11.2 deletion syndrome. However, recurrent candidiasis and/or severe viral infections indicative of clinically significant T lymphocyte dysfunction are uncommon, and complete lack of T lymphocytes associated with a severe combined immunodeficiency phenotype occurs in fewer than 1% of these patients.1–3 The heterogeneous nature of this clinical syndrome makes diagnosis and prediction of the natural history a challenge.

Without measuring in vitro parameters of immunity, it is currently not possible to accurately predict which patients with 22q11.2 syndrome have T lymphocyte dysfunction.4 Assays of T lymphocyte function are labour intensive and expensive. One aim of this current study was to determine whether in addition to assessment of absolute T lymphocyte numbers, in vitro measurement of T lymphocyte function (specifically phytohaemagglutinin (PHA) mitogen responses) is a necessary part of the investigation of these patients, particularly as a previous study of 45 patients with DiGeorge syndrome has shown that the result is usually normal.5–8 A second aim of this study was to determine whether the presence of specific non-immune/infection-related clinical features predicted the likelihood of T lymphocyte dysfunction in patients with 22q11.2 deletion syndrome.



Patients with 22q11.2 deletion syndrome were identified from laboratory and clinical databases within the two regional clinical immunology centres in the UK. Inclusion criteria were any patient with (a) a 22q11.2 microdeletion, (b) available hospital records providing clinical details, and (c) details of assessment of T lymphocyte immunity by lymphocyte subset analysis and PHA mitogen responses. Clinical features were retrospectively collated from the review of the case notes. Adjusted plasma calcium and parathyroid hormone results were also noted where available. For the purposes of this study, patients were labelled as hypoparathyroid if they had (a) biochemical evidence of hypocalcaemia with or without low serum parathormone concentrations, (b) neonatal tetany or noctural muscle cramps that responded to calcium supplements, (c) nephrocalcinosis on renal ultrasound scan, or (d) severe enamel hypoplasia.

As this was a retrospective study not involving any direct patient contact and where clinical and laboratory records were entered into an anonymised database, discussions with the local research ethics committee indicated that a formal ethics approval was not required.

Measurement of T lymphocyte numbers using flow cytometry

Lymphocyte subsets were measured by flow cytometry using standard methodology on one occasion, usually at the time of diagnosis. All blood samples were taken at a time when patients were free from major intercurrent illness that may have affected test results.

Assessment of T lymphocyte function by PHA-induced mitogenesis

Lymphocyte transformation responses to PHA were carried out using standard methods.9 This involved measurement of maximal incorporation of tritiated thymidine into stimulated peripheral blood mononuclear cells after 4 days of culture.

Statistical analysis

Data were analysed with SPSS 15.0 software package (SPSS Inc). As much of the data were not normally distributed, non-parametric statistics were used throughout. χ2 Statistic was used for discrete data. Kruskal–Wallis H (groups of three or more) and Mann–Whitney U tests (comparison of two groups) were used for continuous data. Relative risk (95% CI) for continuous variables was calculated using Cox regression analysis with ‘age at testing’ as a covariate. Correlation between continuous variables was assessed using Spearman's rho test. Data were considered statistical significant if p<0.05 (two-tailed test).


Demographic and clinical characteristics of the study cohort

Sixty-four (74%) of the 87 patients with 22q11.2 deletion syndrome identified between 1993 and 2009 for whom there was a complete data set were included in the analysis. Clinical features of the cohort are summarised in table 1. For the purpose of analysis, patients were divided into three groups based on infection history. Thirty one (48%) had no significant problems with infection, 21 (33%) had a history of recurrent bacterial ear (>3 in a 12-month period) and/or chest infections (>1 in a 12-month period) requiring antibiotics but no history of thrush or severe viral infections, while 12 (19%) had a history of recurrent thrush requiring antifungal treatment (nine patients) and/or viral bronchiolitis and enteritis (six patients) requiring hospital admissions (see table 2 for further details). Samples for virology were not collected from patients who did not have clinical evidence of viral bronchiolitis or enteritis.

Table 1

Clinical and biochemical features of 64 children with 22q11.2 microdeletions

Table 2

Twelve 22q11.2 deletion syndrome patients with severe or recurrent fungal and viral infections

Nineteen of 57 patients for whom data were available had evidence of hypoparathyroidism with hypocalcaemia (table 1). One suffered from neonatal seizures, three from night-time muscle cramps that improved after bed-time calcium supplementation, one had evidence of nephrocalcinosis on renal ultrasound, and one had enamel hypoplasia. Nine had persistently low plasma calcium concentrations requiring calcium supplementation.

Correlation between infection history and T lymphocyte numbers/function

In 30 (47%) patients, the immunology blood tests were taken at or before the age of 12 months. Based on published normal ranges,10 39 (61%) patients had low absolute CD3+ T cells numbers (two with 100 CD3+ T cells/μl indicative of a severe combined immunodeficiency), but only 4 (6%) had lower than normal PHA mitogen responses.

Absolute number of lymphocytes (T, B and natural killer (NK) cells) and PHA mitogen responses of children with a history of bacterial infections, but no problems with thrush or viral bronchiolitis/enteritis, were not significantly different to children with no problems with infections (figure 1). In contrast, compared with children who had no significant infection history, children with a history of recurrent thrush or viral bronchiolitis and/or enteritis requiring hospital admission had significantly lower mean (interquartile range); relative risk (95% CI) absolute CD4+ (475 (440–860) vs 1079 (620–1577); 0.3 (0.1 to 0.8)) and CD8+ (288 (230–375) vs 517 (425–728); 0.3 (0.1 to 0.7)) T lymphocyte numbers, as well as lower PHA mitogen responses (13 602 (505–28 066) vs 31 775 (25 094–51 204); 0.4 (0.2 to 0.8)) (see figure 1). CD19+ B lymphocyte numbers were significantly higher in this group (1987 (838–2218) vs 663 (272–870); 4.6 (1.2 to 16.9)). CD56+ NK cell numbers were similar in all groups. Differences for absolute lymphocyte numbers and PHA mitogen responses were independent of the age at immune testing, determined by using this variable and clinical infection history as a covariate in multivariate Cox regression analysis.

Figure 1

Box plots showing the association between infection history and in vitro parameters of lymphocyte numbers and T cell function. Panels show absolute blood CD3, CD4, CD8, CD19, CD56 concentrations (cells/μl) and phytohaemagglutinin (PHA) mitogen responses (cpm) for patients grouped based on whether they had (a) no significant infections (white boxes) (n=26), (b) bacterial infections only (light grey boxes) (n=19), and (c) additional thrush and/or viral bronchiolitis/gastroenteritis (dark grey boxes) (n=12). Boxes represent interquartile range with the thick black line representing the median. *p Value <0.05 is for multivariate Cox regression analysis with control (no significant infections) as reference group and ‘age of test’ as covariant.

Correlation between infection history and other clinical parameters

The possible association between infection history (recurrent candidiasis and/or severe viral infections (bronchiolitis or gastroenteritis)) indicative of suboptimal T cell function and other clinical parameters was then examined. Patients with a history of these infections were significantly younger (median (interquartile range) of patients with no infections was 10 (4–13) years, for those with bacterial infections it was 7 (3–10) years, and for patients with additional fungal and/or viral bronchiolitis/gastroenteritis requiring hospital admission it was 4 (1–6) years; p<0.01). There were no significant differences in the genders of these groups. Although there was no association between these infections and a history of congenital heart disease or palatopharyngeal anomalies (diagnosed after assessment by ear, nose and throat specialists), all 12 patients with recurrent fungal/severe viral infections had features of hypoparathyroidism, while this was only the case in 1/24 (5%) of patients with no infection problems and 6/14 (32%) of patients with bacterial but no fungal/severe viral infections (p<0.0001) (figure 2).

Figure 2

Histogram showing association between clinical parameters and infection history. White bars are patients without significant infections, light grey bars are those with bacterial infections only, and dark grey bars are patients with a history of thrush and/or viral bronchiolitis/gastroenteritis requiring hospital admission. Numbers in the subgroup are given inside the bars. **p Value <0.0001 derived from χ2 analysis.

Correlation between immune function and clinical phenotype

In view of the strong association between fungal/viral infections and features of hypoparathyroidism, the possible association between T cell function and this clinical parameter was also investigated (table 3). Compared with patients who had no infection problems, those with fungal/serious viral infections had significantly lower absolute CD3+ T lymphocytes (0.5 (0.3–0.9)) and PHA mitogen responses (0.4 (0.2–0.7)), and higher CD19+ B lymphocytes (2.6 (1.0–7.0)). This association was independent of the age at immune testing, as assessed using multivariate analysis. There were also no significant differences between laboratory measures of immune function and presence or absence of congenital heart disease and palatopharyngeal anomalies.

Table 3

Association between clinical and immune parameters


This study is believed to be the first to demonstrate the strong association between clinical features of thymic and parathyroid dysfunction in patients with 22q11.2 deletion syndrome. Patients with clinical/subclinical hypoparathyroidism had significantly lower absolute T lymphocyte numbers and PHA mitogen responses, as well as higher rates of recurrent candidiasis and severe viral infections such as bronchiolitis or gastroenteritis. This is in contrast to the lack of correlation between thymic dysfunction and congenital heart disease or palatopharygeal anomalies.

The reasons why this specific association has not been reported previously is probably twofold. First, the numbers of patients in previously published cohorts may have been too small to provide the statistical power necessary to adequately test for this association.4 6 7 Second, this study has focused on infections likely to be specifically associated with a T lymphocyte immunodeficiency, namely propensity to recurrent fungal and severe viral infections, rather than grouping all bacterial, viral and fungal infections together. In 22q11.2 deletion syndrome, bacterial respiratory tract infections might relate not only to abnormalities in T lymphocyte immunity, but also to structural and functional palatopharyngeal anomalies. We specifically looked for an association between palatopharyngeal anomalies and bacterial ear and chest infections but none was found (p=0.3).

The association between clinically significant thymic and parathyroid dysfunction can be explained by the underlying tissue field defect in the syndrome and the ontogeny of these two tissues. Components of both tissues are derived from the third and fourth pharyngeal pouch structures and thus gene defects leading to the maldevelopment of these structures would be expected to impact on both thymic and parathyroid function.11 12 In contrast, the primary palate is derived from the innermost part of the intermaxillary segment of upper jaw. Likewise, conotruncal cardiac structures are not derived from the pharyngeal pouches but from adjoining branchial arch arteries, an anatomically distinct entity, and therefore defects in genes coding for their development would be less likely to impact on the thymus and parathyroid glands.

Measurement of PHA mitogen responses added little to the measurement of absolute T cell numbers, which appeared to be a more sensitive measure of thymic dysfunction. In this study, PHA responses were only low in patients with almost complete lack of T lymphocytes. Low T lymphocyte numbers in the syndrome relate to a decrease in CD4+ and CD8+ cells, as would be expected from a defect in maturation in the thymus. Another novel finding in this study was the increased numbers of B lymphocytes in the subgroup of patients with poorest T cell function. 22q11.2 deletion syndrome is associated with an increase in autoimmune phenomena (eg, autoimmune cytopenias caused by autoantibody production) particularly in older children and adults.1 Serum immunoglobulin concentrations and autoantibody profiles were not available in this cohort. Prospective studies may shed some more light on the association between subnormal T lymphocyte function and aberrant B lymphocyte activity. It is possible that this may relate to reduced numbers of regulatory T lymphocytes in some patients.13

There are a number of limitations of this study. Its retrospective nature means that more direct laboratory measures of thymic function, for instance in terms of thymic emigrants, T cell receptor excision circles or Vβ repertoire, were not examined, neither was more detailed phenotypic classification of clinical features such as palatopharyngeal dysfunction, which would have been interesting in relation to its effects on bacterial respiratory infections. More detailed, prospective studies specifically focussing on re-evaluating some of the major findings of this study are required to better understand the genotype–phenotype variations in this syndrome. Finally, previous studies suggest that the thymus undergoes maturation after infancy in some patients with 22q11.2 deletion syndrome, and therefore T lymphocyte numbers and function taken in infancy may not reflect these parameters later in life.8 In order to check for potential confounding effect of age, ‘age of testing’ was added as a covariate into the Cox regression analysis. This covariate did not affect the significance of the observed associations.

In summary, we have shown, for what is believed to be the first time, that in patients with 22q11.2 microdeletions, parathyroid insufficiency correlates strongly with low absolute T lymphocyte numbers and an increased risk of severe viral/fungal infections. Low absolute T cell numbers and PHA mitogen responses predict troublesome candidiasis and severe viral infections in this cohort. Measurement of PHA mitogen responses added little prognostic information.

Take-home messages

  • 22q11.2 patients with transient or persistent features of hypothyroidism are more likely to also have reduced T cell numbers and phytohaemagglutinin mitogen responses.

  • This subgroup of patients is also more likely to suffer from recurrent or severe viral infections (eg, thrush, viral bronchiolitis and gastroenteritis).

  • In general, measurement of absolute T cell numbers provides sufficient laboratory information to correctly assess and manage the immunodeficiency of patients with 22q11.2 syndrome; costly measurement of phytohaemagglutinin mitogen responses is usually not required.


View Abstract


  • Funding No external funding.

  • Competing interests None.

  • Provenance and peer review Not commissioned; externally peer reviewed.

Request Permissions

If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.